MESI Caching work start (was Re: caching and userapi)

Matthew Dillon dillon at apollo.backplane.com
Sat Jul 19 15:01:58 PDT 2003


:On Sat, 19 Jul 2003, Matt Dillon wrote:
:
:>     Interesting.  That article is dated 4 Dec 2001.  I wonder if Larry
:>     has updated his thinking since then.  I also looked at the slides
:>     and flipped through the two papers mentioned.  He doesn't specifically
:>     mention MESI but it is obvious that he is angling for the same effect.
:
:MESI would well for caching the metadata of the / directory
:between the OS instances (except for atime updates), but I'm
:not sure it would be a good strategy for eg. /tmp ...
:
:Rik
:-- 
:Engineers don't grow up, they grow sideways.
:http://www.surriel.com/		http://kernelnewbies.org/

    Yes, there are some interesting tradeoffs between a shared caching model
    and a per-cpu caching model.  The book is completely open in regards to
    which model we use for what cache.  I actually don't mind the
    read-copy-update idea, I would just like to find a more efficient way
    to handle updates.

					-Matt
					Matthew Dillon 
					<dillon at xxxxxxxxxxxxx>





More information about the Kernel mailing list